The Fresh Loaf

A Community of Amateur Bakers and Artisan Bread Enthusiasts.

My Tangzhong Roux FAQ

Antilope's picture
Antilope

My Tangzhong Roux FAQ

I put together this FAQ about using a Tangzhong Roux in yeast breads based on my experiences using the technique.
.
.
.
--WHAT IS A TANGZHONG ROUX?--
.
A Tangzhong Roux (also called a Tangzhong Water Roux or Water Roux) is a flour and water roux that is added to yeast bread recipes. This is done in order to make a loaf of bread that is lighter, that has a more tender crumb and a longer shelf life.
.
Bread flour or all-purpose flour is usually used to make the roux. Water is the liquid usually used to make the roux, but milk or a mixture of milk and water can also be used, if desired.
.
The flour and water are mixed and heated to 149-F (65C). This gelatinizes the flour and forms an unflavored translucent pudding-like roux. The roux ingredients come from the original recipe amounts. This roux is added to remaining liquids in the yeast bread recipe. The water roux traps and retains moisture during baking. Using this technique is similar to adding pudding to a pudding cake. The final result is a moister, lighter loaf of bread with a more tender crumb and a longer shelf life. These beneficial effects are all the result of the moisture retained by the water roux during baking.
.
The Tangzhong roux technique was developed in Asia around 2000. The technique was first mentioned by Yvonne Chen in her book, “Bread Doctor”, published in Taiwan in about 2003. Tangzhong means "soup" in Chinese.
.
.
--WHAT YEAST BREAD RECIPES BENEFIT FROM A TANGZHONG ROUX?--
.
The Tangzhong roux technique will work with pretty much any yeast bread recipe, making a lighter, more tender and longer lasting loaf of bread. I've only used the technique for straight dough breads. I haven't used it for sponge breads, etc. But there is no reason it shouldn't work on those, also.
.
I've used the Tangzhong roux technique on white bread, sourdough bread, hawaiian bread, cinnamon rolls and cinnamon swirl bread, light wheat bread (part bread flour and part whole wheat flour), rye bread, vienna bread, French bread, hamburger buns, lean breads, rich breads and sweet bread doughs, etc.
.
The Tangzhong roux technique will work for most hand kneaded, mixer kneaded and bread machine recipes. I've even used it on 65% hydration, stretch and fold, yeast bread recipes.
.
The only type of yeast bread recipe where it didn't seem to have much effect was one that used whole wheat flour to make the water roux for a 100% whole wheat bread. It didn't seem to lighten the loaf very much. However, using 3 Tablespoons (20g) of white flour in the roux of the 100% whole wheat loaf did seem to lighten it. In this case, we are adding 3 Tbsp (20g) of white flour to the recipe, so remove 3 Tbsp (20g) of the whole wheat flour to keep the recipe in balance.
.
When I make a light wheat bread (part bread flour and part whole wheat) I take 3 Tbsp (20g) of the roux flour from the white flour. This will lighten the light wheat loaf, don't use whole wheat flour in the roux.
.
.
--HOW DO I USE A TANGZHONG ROUX IN AN EXISTING YEAST BREAD RECIPE?--
.
Use 3 level tablespoons (20g) of white flour in 1/2 cup (120g) of water, for either 1-1/2 lb (750g) or 2 lb (1kg) yeast bread loaves. I use bread flour or all-purpose flour in the roux.
.
The Tangzhong roux ingredients come from the original recipe amounts. Don't add extra amounts. Measure out the original recipe ingredients and take the Tangzhong roux ingredients from that. If the recipe uses only milk, make the roux using milk.
.
I make the roux in a microwave. Mix 3 Tbsp (20g) of flour and 1/2 cup (120g) of water (or milk) in a microwaveable cup. Microwave on High for 25 seconds. Stir well. Microwave 15 seconds more. Stir. The roux temperature should be at about 149 F (65C) and a white translucent pudding should have formed. If not, microwave another 5 seconds and stir well. If necessary, continue microwaving 5 seconds at a time and then stir well, until the white translucent pudding forms. I use a 1000-watt microwave.
.
The roux can also be made on the stovetop in a saucepan.
.
Be careful to only heat the roux to around 149-F (65C). Five or ten degrees more doesn't hurt. I haven't explored heating the roux higher than that. I'm not sure what higher temperatures would do to its effectiveness. References I have reviewed don't address overheating the roux.
.
I mix the hot roux into the remaining recipe liquid ingredients immediately. Mixing well. The recipe liquid will then end up lukewarm. The roux can also be cooled to room temperature first or stored in the fridge, if desired.
.
Once you confirm that the recipe liquid is lukewarm, it's then safe to mix in instant yeast or proofed active dry yeast, if desired. Otherwise, follow the original recipe instructions for adding yeast. Just be careful to keep the hot roux away from the yeast.
.
After mixing the Tangzhong roux into the remaining recipe liquid ingredients, continue with your original yeast bread recipe just as you always do.
.
.
--HOW MUCH TANGZHONG ROUX SHOULD BE USED IN A YEAST BREAD RECIPE?
.
I use 3 level tablespoons (20g) of white flour (bread or all-purpose) in 1/2 cup (120g) of water, for either a 1-1/2 lb (750g) or a 2 lb (1kg) loaf.
.
Here is how these quantities are arrived at: (You really don't have to be this exact, unless you want to, this info is just for background info):
.
Most Tangzhong roux references say to use 5% of the total recipe flour weight as roux flour. The roux flour is mixed with five times its weight in water. So a loaf using 500 grams of flour would use 25 grams as roux flour (5% the total flour weight) and 125 grams of roux water (5 times 25 grams ). Both of these ingredient amounts are taken from the original recipe ingredients. Extra amounts are not added to the recipe.
.
Once again, this is roughly equal to about 3 level Tablespoons (20g) of flour mixed in 1/2 cup (120g) of water. I use this same amount in 1-1/2 lb (750g) and 2 lb (1kg) loaves. It's close enough for the recipes I have tested.

GregS's picture
GregS

Thanks Antilope. This general guidance will encourage me to expand the Tangzhong method into some of my other yeasted recipes. Clear and concise.

GregS

clazar123's picture
clazar123

The purpose behind using a tangzhong is to increase the starchy gel in the dough, which is part of the matrix of a good crumb. I have often successfully used the technique with WW bread using a portion of the WW flour and most/all the liquid. However, WW being what it is, the tangzhong will only be helpful if the basic requirements of WW are adhered to. That is-adequate hydration, some form of dough rest so all the bran can absorb the extra water and kneading to windowpane to fully develop not only the gluten but the starchy gel.

Good write-up!

jkandell's picture
jkandell

"The Tangzhong roux technique was developed in Asia around 2000." I think the method of gelatinizing some (5-25%) of the flour was developed many decades earlier.  I know Auerman wrote about it with Russian rye in 1935, and I think the early soviet bread writers were just codifying pre-soviet methods originally from frisian breads.Lots of  traditional Scandinavian, German, Russian breads use scalded flour. The difference was they didn't use this method with white flour the way Asians do. 

Mini Oven's picture
Mini Oven

I found, over heating the roux mixture reduces hydration to the point of having to replace some of the evaporated water in the recipe.  My tip is to weigh the flour, water, and vessel (dish or pan) before heating and then again once cooled.  Replace any missing weight with water before mixing into dough.

Mini

valereee's picture
valereee

So smart to weigh the pan before and after!  I am totally adding that to my normal procedure when using a tangzhong!  

Antilope's picture
Antilope

it doesn't even boil, so there is no moisture loss. It's done in 45 seconds. In a saucepan on the stove top, yes there can be moisture loss.

valereee's picture
valereee

Will/should my just-mixed, unkneaded/unfermented dough start out feeling/looking slightly drier than normal when I've used a tangzhong?  

Mini Oven's picture
Mini Oven

:)  That is why I tend to use it with high hydration doughs, to get some control over the dough.  The dough with tangzhong will behave itself more and be easier to shape.

valereee's picture
valereee

Mini Oven, the no-knead bread is very high hydration.  But I just used a 12% tangshong with my regular recipe, and it mixed up drier than usual.  I was wondering if that meant I should add a bit more water.  

 

Mini Oven's picture
Mini Oven

If the hydration of the water roux stayed the same (same weight as before heating) then yes, the dough will feel dryer.  I would not add more water for this is normal.   Unless you want to raise the overall hydration of the loaf.  (could result in shorter fermenting time and longer baking time)

12% flour in the tangzhong is rather high.  You may find that the finished crumb has not changed at all or gotten heavier.

valereee's picture
valereee

Thanks, Mini Oven!  I didn't add more water, but it was still a bit dry the next morning after fermenting.  It rose okay but didn't get much oven spring.  I'm thinking maybe I boiled off a bit of the water, as I hadn't weighed.  

not.a.crumb.left's picture
not.a.crumb.left

Hi Mini, I know this is an old thread but I am intrigued by this method. Let's say if I make a 500g SD loaf how much water and flour would you use for the roux and could I add it after the autolyse with salt and levain and still use Rubaud method? 

I would like to experiment with this to manage wetter doughs better and ideally not affect the openness on the crumb. I hope this makes sense.  Kat

Doc.Dough's picture
Doc.Dough

As Mini points out, the normal roux content is made up of ~5% of the total flour.  So if your 500g sourdough loaf is made at 67% hydration you have about 300g of flour and 5% of that would be 15g.  5X flour weight of water would be 75g which would yield something like 80g of roux after all of the losses.  But I would go the other way and make a larger batch of roux (maybe 20g flour + 100g water) and use 90g of it in the dough. I think it is hard to make a very small batch of roux so you may want to double that.  It will account for 15g out of the total 300g flour and 75g of the 200g of water.  Of course you will account for your starter in the total in your usual way.

The Tangzhong method was apparently developed to facilitate a soft tight crumb so a roux will affect the openness of the crumb (at 5%, I would expect the crumb texture to be a lot like commercial sandwich bread).  The flavor will be that of sourdough since the LAB still will produce acid. 

 

not.a.crumb.left's picture
not.a.crumb.left

to put on my list to do!  Kat

Dianukas's picture
Dianukas

Hello Doc. Dough,

You seem very knowledgeable.  I have read a lot if your comments. I would like to ask you a few questions.  I think no one can give me a better answer than you. Its regarding dehydrating culture at its peak from different flours. I couldn't figure out how to send private messages on freshloaf.com. 

much appreciated.  

Sorry everyone for spamming this discussion...

Lechem's picture
Lechem (not verified)

and ask him to enable your private messaging and editing.

floydm@thefreshloaf.com

Dianukas's picture
Dianukas

Tnx a lot for advice 

Lechem's picture
Lechem (not verified)

Edit out your personal email address above and send the original poster a private message.

I'm no expert but I do know how to make a tangzhong so feel free to ask me or anyone else here.

Doc.Dough's picture
Doc.Dough

I would separate the issue of dehydration from the use of different flours.  Your culture is defined by the yeast and LAB and not by the flour. Your starter is defined by the culture and the substrate you choose to feed the culture.

Dianukas's picture
Dianukas

Thank you. I have sent you a private message. 

Calvinize's picture
Calvinize

Hi doc, what would happen if I use a higher percentage of tangzhong to the total flour? Say, instead of 5% of the total flour, I use 10% or 12%. Does the bread get even softer? Or is it the other way around?

Doc.Dough's picture
Doc.Dough

@Calvinize - I don't find that case in my files, so I don't really know.  How about running that case and showing us what results you get?

Doc.Dough's picture
Doc.Dough

I have found that the cell size of a high hydration product made with the tangzhong method is not as large as the same batch made without the roux. As Mini points out, the dough is easier to handle as well but you do make a trade-off to get that benefit.

valereee's picture
valereee

Doc.Dough, I think what you're saying is that the holes in the crumb will be smaller?  

Val <----not clued into the jargon

Doc.Dough's picture
Doc.Dough

Bread is a foam so the elementary units are cells separated by gas-tight sheets of protein. You got it.

The first time I made a batch of 75% hydration dough with 20% of the flour in a roux I was amazed at how much easier it was to handle.  I had never been able to successfully slash a dough that wet, but sure enough it was easy to do.  However, when it came out of the oven the crumb texture looked a lot more like a commercial hamburger bun than a ciabattini.  The taste was fine.

Now I am wondering if the water that boiled off in the making of the roux was enough to alter the dough texture.  I really like Mini's process enhancement to force it back to the right water weight.  I will have to try it again and see what happens (next time I need some hamburger buns).

Thank you (again) Mini!!

David Esq.'s picture
David Esq.

Mini says that the process generally callAs for 5%, and notes that 12% is high. What is the reason behind your using 20%?  I am not sure the trade off you noticed is relevant when using 5% of the flour in the roux. 

Doc.Dough's picture
Doc.Dough

@David - Good catch.  I went back to the data sheet and the roux was made with 50g of flour and 250g water (this is the 20%), but the bread was made with half a batch of roux (25g of flour) plus 350g of flour, so the batch had 6.6% of the total flour in the form of roux.  It also had 8% solid fat in the mix so that may have contributed to the tight crumb.

And a batch of 250g water + 50g flour heated for 2:50 on Medium loses ~4g of water to evaporation and 2g to the whisk (and then you determine how well you clean the pan to yield something close to 290g of roux.

Avibabyau's picture
Avibabyau

Hi. This process is very new to me so I have questions. I usually make sour dough with a starter. Does the roux replace the starter? Or is it an adjunct to produce a softer crumb only?

Doc.Dough's picture
Doc.Dough

When I make sourdough tangzhong rolls the roux replaces the corresponding amounts of flour and water and I use the regular amount of starter/levain.

Lechem's picture
Lechem (not verified)

Is a bread improver. Not a starter. You'll still need your starter.

Mini Oven's picture
Mini Oven

Einkorn tangzhong bennefits greatly when fully hydrated before heating.  Allow at least 30 minutes wait to avoid a gritty mouth feel.

maryjobo's picture
maryjobo

According to PJ Hamel at King Arthur Flour (https://blog.kingarthurflour.com/2018/07/23/how-to-convert-a-bread-recipe-to-tangzhong/) you have to first convert your recipe to 75% hydration.  You can't just take a recipe that's say, 65% hydration and use the existing amounts of flour and water to use in the tangzhong.  She shows a side-by-side comparison of a loaf made from an original recipe as a control, one loaf of the recipe converted to tangzhong without changing the hydration, and a loaf made starting from the same recipe which is then converted to 75% hydration before calculating the tangzhong amounts. Pretty cool photos.

Mini Oven's picture
Mini Oven

going on in the test recipe using milk.  Milk is not water, it contains roughly 10% milk solids and 0 to 3.6% fat so the hydration will be much lower.... water divided by flour figures hydration.  So when figuring the amount of milk used to make the Tangzhong, there will be an increase. One has to add between 10 to 13.6 % more milk to get the correct hydration that water would give.  This was not mentioned.  Actually the hydration of the recipe is lower than the calculation.  

Doc.Dough's picture
Doc.Dough

The photos are indeed interesting, but I was looking for the full test matrix to include the basic formula modified to 75% hydration without any tangzhong, and that data point was missing. I also find it curious that the baker was unable to round up exhibit B into a boule.  Even bagel dough at 55-60% can be formed into a ball.

You do have to account for the fat in the milk (as Mini points out) but you also have to account for the loss of water that evaporates when you make the roux.  And for small batch sizes that can be significant.

Interesting article, but I am not convinced by the rationale given that it is necessary to convert the formulation to 75% hydration. The basic recipe may be just fine at 75% hydration without any tangzhong, and that is the formula that should be the basis for comparison, and when the conversion to use tangzhong is done, the details have to be handled with care so that the correct amount of water is added (or make extra roux and use an amount that is appropriate to make up for the evaporation of water and the milk solids and fat fraction of the milk).

Pompom's picture
Pompom

As much as the tangzhong helps with bread texture and moisture, I was wondering if anyone has used this with a cake recipe?  Wouldn't that make a cake that much more moist and tender too or would it make it a bit more dense?

Pamela

Doc.Dough's picture
Doc.Dough

I await your results.  But you have to do two cakes at the same time with one of them using the pre-gelatinized starch.

Mini Oven's picture
Mini Oven

as a small amount of starch is used to thicken milk and then the pudding is added to the cake batter.  :)

Maybe we should call Tangzhong breads, Pudding breads or Porridge breads.  :)

trijezdci's picture
trijezdci

I am sorry but this technique has not been invented in Asia, and certainly not in 2000.

It has been in use all over Europe for centuries.

In fact, it may go back as far as the medieval ages. Flour-water roux has traditionally been used in spelt bread because spelt bread tends to dry out during baking. And spelt bread was very common in the medieval ages in Central Europe.

The technique is also used in rye breads that contain whole grain. One prominent example is Borodinsky Chleb of Russia, for which the use of a flour-water roux is documented in a recipe of 1933.

In Austria, Germany and Switzerland, a flour-water roux is just one variety of roux used in bread dough, and collectively they are known as Nullteige (lit. null doughs or zero doughs). There are three categories:

(1) Quellstück (lit. swell or soak pieces)

flour, grist, grain or seeds are soaked in cold water (or water of ambient temperature) for up to 24 hours, to be added to a dough the following day

(2) Brühstück (lit. brew piece)

flour, grist, grain or seeds are soaked in hot water for one to six hours, to be added to a dough the following day

(3) Kochstück (lit. boil piece)

flour, grist, grain or seeds are soaked in water and then boiled into a roux, to be added to a dough the following day

The flour-water rough described in the article belongs into the latter category.

This is part of a baker's curriculum during their three year's apprenticeship in Austria, Germany and Switzerland and has been for at least a century. It is likely part of a baker's training/education in other European countries, too.

Somebody in Asia may well have re-discovered the technique independently, but to say it has been developed there as recently as the 2000s, is a gross misrepresentation.

Doc.Dough's picture
Doc.Dough

Thank you for both the history and the vocabulary lessons.

As is often the case, the inventor gets little or (more often) no credit, and the advocate who shouts loudest appears to be the winner.

The Roadside Pie King's picture
The Roadside Pi...

Ellie's picture
Ellie
  • hydration is bang on at 75% not including evaporated milk in my calculations, but hydration is at 81% once included? If you say evap milk should be included, do I reduce my hydration back down to %75, reducing the water in the ingredients? Thanks
Doc.Dough's picture
Doc.Dough

Include the evaporated milk in the liquid column when you calculate the hydration, but remember that hydration is an inexact term since there is more to bread than flour and water.  It gives you an idea of what the dough texture will be but does not predict the results of your bake.

Ellie's picture
Ellie

Thank you. Will experiment again today with evap milk included and with higher protein flour. Yesterday’s bread wasn’t a good result. Really soft but still crumbly. 

Mini Oven's picture
Mini Oven

add milk solids to the flour.   Dont forget milk contains a fair amount of fat.  Might have to adjust fat amount as fat can give a cakelike crumb.

Ellie's picture
Ellie

That’s good to know. Thanks

tilerdon's picture
tilerdon

I only just came across this a few weeks ago on a Sourdough Forum and after a little research decided first of all to apply it to my yeasted sandwich bread which I was baking 3 times a week but which was only better either toasted or with a relatively wet filling by the 3rd day.

The articles I homed in on at first all seemed to point to increasing the hydration of the dough to 75% and that this is what TangZhong is really about rather than just incorporating a water roux into the mix. The results have been pretty successful though I am not getting the oven spring that I expected. 

There are obviously many things that I can experiment with - the percentage of the flour that I use in the roux and of course the % hydration of the dough.

I've designed a spreadsheet that lets me vary these factors and produces an instant result so there is no effort in the calculations. It also allows me to use other liquids that don't have 100% water content and sourdough starters of any hydration level for sourdough recipes.

So I'd love to hear suggestions as to which direction I should head

Rick Reid's picture
Rick Reid

Bakers are disputing the 5/1 ratio.
They are discovering that a ratio of 10% - 40% is ideal, depending on how soft you want your bread to be.

Rick Reid's picture
Rick Reid