I've heard (and experienced first hand) that cake flour will pack tight in the hand when squeezed and form basically a small corn cob shape. Colonel Sanders describes this flour when he describes the proper flour for KFC. I did this test with the industry standard (Dawn) donut mix and the same result happens (packs tightly). However, it must not be the flour itself causing the tight packing effect, but rather the powdered egg, milk, sugars, etc. I tried my usual donut recipe with cake flour and I'm here to say, that cake flour is not the answer. I tried GM Purasnow flour which is around 8.2% protein and gluten development is simply not going to happen. With this said, Dawn mix most likely is not cake flour, or if it contains cake flour, it's within a certain ratio that still allows for good gluten development.
Why try cake flour in donuts? Because although we want gluten development, we also want soft and airy. I consider Dawn perhaps the most expert in the donut industry based on the fact that it's what most donut shops use commercially. They also make the mix for Krispe Kreme. 100 years of experience as a top donut ingredient/mix supplier and a huge staff most likely (or perhaps certainly) give them more experience than anyone else in the world. So with this said, I figure that they probably have the perfect blend of flour figured out. Is there a way to reverse engineer (roughly) the flour composition without a lab analysis?
Why not just use Dawn mix? Well personally, I like the taste of a fresh homemade donut made with real egg, milk, fat, etc better than the Dawn mix. But... I don't like the chew. I want mix tenderness with homemade flavor, if somehow the best of both worlds were possible.
Which type of donuts do you mean - yeast-raised, cake, or what? For soft and tender, I would try White Lily All Purpose, which is bleached, finely milled, and low in protein or their self-rising flour, if you are going to use baking powder for the leavening agent. Or blend pastry (instead of cake, which can be finer in texture) and AP flours and see if that's better.
TomP
Yes, important detail I missed there! This is for yeast raised donuts.
Start with something easy - then tweak. Eventually you'll get it. Enjoy!
The thing about ultra processed foods is that they use ingredients and processes that you aren't going to have reasonable access to as a home cook and they'll have invested more into R&D than you're ever going to be able to. All for the purpose of creating a hyper palatable product that you can't get anywhere else.
So you might not be able to get the exact texture you want in a homemade donut. That said, looking at the ingredients list for Dawn's Raised A donut mix (which seems to be their mix for yeasted donuts) does point to at least one thing you can easily do to get an airier texture.
https://www.walnutcreekfoods.com/products/baking-supplies-bulk/mixes-ingredients/donut-mix-raised
Add chemical leavening. In addition to the yeast.
I don't have a thorough understanding of the chemistry of quick breads but I think you want double acting baking powder because it's activated first by liquid and later by heat; I believe baking soda does not get the second rise from heat. Since you're also using yeast, you might lose the air from the liquid rise by the time you get to cooking the donuts. Handling technique may be able to reduce that loss? I don't know enough about working with chemical leavening to say.
I suspect this is going to have a greater effect than using marginally different flours. So perhaps try that first, then if you still aren't satisfied try experimenting with the flour.
This is something I also noticed. Krispe Kreme (made by Dawn) also goes this route. I've tried it over the past couple years on and off and still use baking powder in my main recipe based on principle. I'm not at all sure how much effect it has since proofing is done in a home environment and not exactly scientific. Good vs bad proofing even in the same batch of dough has a huge effect (the biggest difference). So it's not easy to tell what's going on exactly. One thing that is easy to measure is gluten development. Another thing that's easy to notice is texture after frying. This is why I'm experimenting with flour. Texture can be a little tough to tear apart (even though a very good donut overall). I figured Cake Flour would give a more tender product that falls apart, but so far, no luck with that. It just doesn't ever get to window pane.
Getting a window pane isn't essential for making good bread - of course, it depends on the style of the bread - and I doubt it is for donuts, either. In fact, if you are seeking softer more brittle crumb, getting a window pane will work against you. Biscuits (US), scones, and other tender products use very little gluten development.
You would think that window pane would work against the donut baker, yet every set of professional baker instructions I've seen for yeast raised says to machine mix 8-12 minutes or until window pane is reached. I think the reasoning behind this is to make sure the air is trapped well enough for a good rise.
When I tried this with the cake flour, the gluten was such that the dough just tore apart with a jagged edge and no stretch. We rose some of these just for kicks and the result was pretty flat and tough since we lost all of our air. The high gluten product had a much airier internal texture... just a little chewy. So in reality, the donut baker, I believe, is somehow after the best of both worlds gluten wise. There's probably going to have to be a compromise. This ultra tenderness is not such a concern in the bread world (at least not in my bread world... lol).
I was a doughnut-head. I've fried hundreds of batches, I nearly open a doughnut shop a decade ago because I accidentally was too specialised in it so why not (lol). When it comes to doughnut, I go with the Asian way of making fluffy doughnut: high hydration (with tangzhong, yudane, whatever), high gluten content, windowpane, and low fat (I never use solid frying fat because I like the mouthfeel of liquid fat, hence less dough fat, so it won't absorb oil as much)
baking powder only necessary if you move your proofed dough by hand, to compensate for human error during handling. You can proof your doughnut on parchment paper, and fry the parchment paper too (and later fish them out)
Jay
Really insightful breakdown, Nick! I’ve wondered the same thing about commercial mixes like Dawn—there’s definitely some flour alchemy going on. I agree, cake flour alone doesn’t cut it for donuts; too tender, not enough structure. My guess is Dawn uses a proprietary blend—maybe something like a low-protein pastry flour base with a bit of bread flour mixed in to allow for just enough gluten. The tight packing could easily come from the sugar, powdered milk, and emulsifiers. Have you tried mixing AP flour with a touch of cornstarch or experimenting with a blend of AP and low-protein flour to strike that soft-but-structured balance?
There is a wide range of dough conditioners that commercial donut mixes can include. They could make a difference in the finished product.
True... I didn't think much on dough conditioners.
I had cornstarch in my ingredients originally and no idea what it did to the finished product but it didn't seem to suffer (more) in the softness dept when I removed it. I started with a very complex mix of ingredients and have since slimmed it down. Too hard to tell what does what when the ingredients list is long.
I'm going to apologize for consulting ChatGPT and pasting in its list, but this is the kind of thing it ought to be good at. Here's its list for commercial yeasted donuts:
I was going to ask if you had tried yudane/tangzhong in a recipe and I see Jay mentioned that he used it in his doughnuts. It's a simple "conditioner" that makes the crumb softer.
We did try this and it didn't seem to help as much as we'd hoped. Still too much chew to be what I'd consider a normal donut. We also tried potato. It was barely noticeable in the flavor. It did make it noticeably softer but still not quite there. We also tried more fat in the dough. All of this hunting has me thinking that it's either the flour combination (even doubting this now after trying a blend of bread and cake flour) or the dough conditioners. Most likely it's the dough conditioners that are the key here.
Could you summarize what you want in terms of texture/crumb/mouth feel? These are fried, aren't they?
Well it's not so much that the texture needs to be out of the ordinary. It's just that what we have now (compared to the ordinary donut shop donut) is pretty tough, especially after it's cooled. Think beef jerky but not quite to that level. I'm used to a donut being pretty easy to pull apart. The donuts I have now are light and airy and look like a good donut from a shop. But they pull apart like a piece of homemade sourdough bread (the crust of the donut is much like the crust of sourdough in toughness).
Bearing in mind that I've never made donuts, this sounds to me like over-kneaded dough - over-kneaded for donuts, that is. Minimal kneading would give you something like a biscuit/scone, so there must be a happy balance. The crust might be caused by too high a cooking temperature.
What % of yudane did you use?
I think my son and I tried tangzong and yudane (Sp on either??) and as far as I remember, we just followed directions off of a video by chainbaker (first time I'd heard of the methods). It's been a while so I can't remember what the percentage was. It seemed like a good method and useful in some cases I'm sure. In this case, still haven't quite found the solution to this. It's not that what we know how to make now is bad. It's just a little chewy, and more so after 20 or more minutes. Eat one the next day with a coffee and you'll definitely need the coffee to help get through the donut.