No Ordinary Dumpling: Dough
No Ordinary Dumpling: Dough
Welcome to our No Ordinary Dumpling series, where our chef Brian Yong details the R&D process of how we here at XCJ developed our current recipe for our signature xiao long bao (小笼包) , or XLB. He will be analyzing the three fundamental components of a soup dumpling: the dough, the filling, and the jelly (or the soup). In these posts, he’ll share both the highlights — and the struggles — of his development to create a dumpling that we are proud of, as well as a scaled-down, at-home-friendly recipe for anyone interested in making a version of our XLB.
For XLB dough, there are some specific considerations when it comes to taste, texture, and mouthfeel. As with all doughs, the first step is to establish situational awareness: what is the purpose of the dough in this setting?
In our opinion, the dumpling skin of an XLB should be:
Elastic: As the dumplings cook, the soup will turn to steam and expand, applying pressure to the wrapper, which needs to stretch to accommodate the expansion
Tender: Although the dough must be strong to accommodate expansion and manipulation, the cooked dumpling skin should be delicate enough to complement the tender meat and clean broth
Workable: In order to wrap these delicate dumplings properly, the dough must be consistent and pliant, so that the texturally and visually appealing pleats can be achieved.
Not all flours are made the same — even those within the same category like “all purpose.” If you take a look at readily available all purpose flours, you may be surprised to find that each brand has slightly different protein content, hovering around 13%. This is very important as the proteins found in flours made from wheat — glutenin and gliadin — form into gluten when mixed with water in a dough. This then creates the matrix structure necessary to support a dough in its final form. (In a yeasted bun, for example, the dough rises when air becomes trapped in the gluten matrix.)
When it comes to commercial flours, there are different flour blends based on the origin of the wheat and seasonality of seeding. When I first started testing dough, I was blending high protein (14%) and all purpose flour without focusing on a specific brand, which resulted in a lot of variance in the final product — even the color of the dough would change per batch. When I discussed this with other chefs, a colleague who was experienced with pizza doughs recommended I try high quality pizza flours as my base; coincidentally, this protein percentage already fell in-between all purpose and high protein.
After a series of many more tests and discoveries, we finally settled on the use of Mondako flour for all our dumpling dough. This flour is milled from a blend of winter wheat (typically higher in protein content) and spring wheat (typically lower in protein content) to meet at a protein percentage ranging from 11.7-12.3%. This blend delivers an extremely consistent result and, just like its creators boast, truly does have a “high tolerance for frozen goods.” (As in, it freezes well and keeps its flavor.)
With our flour chosen, the next major part of the R&D process was water. The two main points here are hydration ratio (how much water to be used in relation to flour) and water temperature. Because I had previously created a recipe for shen jian bao (生煎包) or Sheng Jian Bao dough for our restaurant in Seattle, I decided to start with a similar hydration ratio knowing I would likely need to adjust it. (SJB dough is leavened, resulting in a different flavor and texture that pairs well with their shallow-fried cooking method.)
I worked in small batches, lowering the hydration incrementally by 1% until I reached 39%. At that percentage, the dough was not too wet for the machine to handle, provided a supple texture when steamed, and also held enough structure to be pleated neatly.
wrappers using dyed water so I could tell what was what.
Throughout this process, I was exclusively testing with room temperature water. Although most XLB recipes you find online will recommend working with hot water because it’s easier to work by hand, this develops less gluten structure. Previously I noticed the XLB restaurants in China I had worked in generally used cold or cool water in their hydration ratios. This decision requires the dough to be mixed more, in order to better integrate the water and flour, and activates more gluten development for a firmer, more toothsome final wrapper.
Finally, the last major key ingredient I tested was salt in the XLB dough, focusing on its impact to flavor and in texture. While the salt added to doughs isn't meant to give it a palpably salty taste, its absence is certainly felt during a taste test. Additionally, salt helps strengthen the dough because it delays the gluten protein hydration, resulting in a finer gluten network that can better withstand processes like freezing and delivery.
I started my salinity testing at 0%, working upwards in increments of 0.1% until its final texture felt substantial without becoming too “al dente.” Through this, I landed at 1% for our final dough, which then went on to be tested for scale. I originally started with just 1 batch of dough at a time (which makes about 320 xiao long baos); this was quickly doubled to 2, 4 and 8 batches, at which point the recipe and handling procedure needed to be modified for consistency..
At this point we received some feedback that the dough was falling apart during the steaming process, which required us to revisit our dough process and its effects on our dough strength. It turned out we needed to shorten our window of dough proofing (more on that below). So to date our salt content continues to remain consistent even as we’ve grown past 12 batches of dough per run.
Adjuncts (other ingredients)
There are 3 other common considerations for XLBs beyond flour, water, and salt. These are yeast, oil, and kansui (枧水), or lye water.
Since XLB dough already doesn’t need to be particularly fluffy or airy like SJBs, I conducted 1 experiment with yeast and quickly ruled it out as it did little to improve the dough.
Kansui was a tip from a friend who often eats at Din Tai Fung, and shared some intel that they use kansui to improve the hardiness of their dough. (You can tell a dough has kansui as it takes on a very distinctive sheen after it’s been cooked and set.) As kansui is extremely potent, I started with just 0.5% and even that made a large difference in the final wrapper. After lowering that amount 0.05% at a time, I eventually ruled it out entirely as the dough was turning out almost too hard, and this addition provided no value to taste, when compared to the dough without kansui. (Kansui, however, is quite useful for making noodles.)
The unique sheen that kansui-based
doughs develop, once cooked and set.
Oil is used to increase the elasticity of dough, making it easier to manipulate during the making process. Oil is also useful in that it is insoluble with water (the two don’t mix), so it creates a slight barrier between the moist filling and the dough, keeping more of that XLB soup intact versus being absorbed into the wrapper. However, since oil is a liquid it causes the dough to become soft and looser, this required additional test trials to work with the machine at volume. I ended up using soybean oil (and very neutral tasting oil), and tested starting at 0.5% and eventually landing at 1.2% for a uniform and pliable dough.
All doughs benefit from a resting period, where the finished dough is left alone so the gluten can relax and form into long strands that create a smoother, more supple texture. (This process is also called the autolyse.) Without a proper rest, I found the dough too tough as it went through the machine, in addition to a much more brittle mouthfeel. This is because the dough has not developed enough extensibility, or the ability to stretch without tearing, and thus the gluten strands were being broken as the machine mixed it. (Note that this is different from elasticity, mentioned in the oil section, which is the ability of the dough to resist stretching.)
I began experimenting with resting periods, starting with 1 hour, and eventually found no particularly noticeable difference between 2 and 5 hours. From there, we attempted to batch out our production so we could prepare dough the night before. However, taste tests quickly confirmed that when the dough sat for too long, it relaxed too much and became undesirably soft once steamed. Now, we start production on the dough first thing in the morning at 3am so we can complete the entire dough process in one day.
When I learned from XLB masters in Shanghai, they would tell me that the dough changes everyday based on the weather, so they would adjust based on feel. As part of the resting process, we quickly learned that temperature and humidity drastically affected the best final product. In particular, hot temperatures (above 80°F) would cause the dough to begin fermentation, which we did not want; I remember one instance when I was shuttling ingredients and equipment between our Bellevue restaurant kitchen and our Auburn production kitchen, a test batch of dough sat in my hot car for a few hours — let’s just say it was not tasty!
Since then, we have implemented a climate control system in our production kitchen to ensure it’s consistently circulating cool air throughout the facility. This way, we can control the temperature and stick to one consistent dough process that we know will work every time.
Storing the Dough
Last but certainly not least, the dough storage process. Dough is a living ingredient that will equilibrate to its surroundings, so improper storage ruins even a perfectly-formulated dough. For us, that means doughs need to be wrapped carefully in plastic to rest and autolyse without drying out or forming a “skin” (or a dry outer layer). If this does happen, the outer layer is later re-incorporated into the soft inner dough, causing an unwelcome textural change. This also can cause major mechanical problems: a small piece of dried dough getting stuck inside our machine will create a split seam that ruins the entire dough log (somewhat like when you have that one little puncture in your plastic wrap). This would then mean I had to disassemble the entire machine, remove the dough, reassemble the machine, and start the dough process all over again.
The Dough Recipe
Sitting here writing this, I can genuinely say I feel really proud of where we’ve landed with our XCJ dough. While the search for the perfect dough still continues, I think you’ll be pretty pleased with the result from this home-friendly version of our XLB dough:
- (300g) - All Purpose Flour
- (140g) - 120°F Water
- (2.8g) - Salt
- (14g) - Vegetable Oil
1) Measure 300g of flour in mixing bowl. Add salt, oil and combine.
2) Measure 140g of 120°F water, pour into mixing bowl. Mix together. Once water is incorporated, use hands to knead until dough comes together
3) Wrap in plastic wrap and rest for 30 minutes