The Order of Things

Those who are familiar with my work know that I occasionally go into lengthy explanations in advance of the actual demonstration. Since we have already cultivated the interest in this topic, we will make it happen first, and then explain just WHAT happened and WHY. Let’s get to it.

Volumes of Velvet

As shown in the slide  below, we started with:

  • Flour, 2 Tablespoons/16g.
  • Bouillon, COLD, 2 cups/450ml.
  • Butter, 2 Tablespoons/30g.

Using a blender on pulse, a stick blender, or a wire whip, or even a wooden spoon, lightly combine the flour and butter with ONE THIRD of the volume of stock. This will form a light paste. In the old days, we used to just mix it with our hands. We were not the least bit squeamish.

DO NOT WHIP. Introduce as little friction as possible! Add the rest of the bouillon, and mix lightly. I will explain why later.

Transfer the mixture into a Ziploc Freezer Quart bag. Process in a preheated sous vide bath set at:

183F/84Cx 1 hour.

As part of the normal process, you will see the thicker substance settle to the bottom, and then slowly assimilate with the thinner liquid above. We will occupy ourselves with something else in the meantime. The IC provides enough agitation, and we all know that the great thing about sous vide is that you don’t have to stand there and watch it!


Let’s do it the old school way, so we have some comparison:

  • Flour, 2 Tablespoons/16g.
  • Bouillon, COLD, 2 cups/450ml.
  • Butter, 2 Tablespoons/30g.

Heat the stock to 183F/84C or thereabouts. I used the microwave. Yes, that’s right, the red headed stepchild is capable of heating water.

Melt the butter in a small sauce pan, add the flour to make a roux, but do not cook more than a minute–we don’t want the roux to contribute any color to this sauce.

Turn the burner to low, and add ONE THIRD of the heated bouillon. This will cause the sauce to become thick IMMEDIATELY. Add the rest of the bouillon, and finish incorporating. Again, DO NOT WHIP. Again, I will explain why not later. Allow to simmer for a few minutes, and set aside. If you are anxious about lumps, you CAN strain at this point, but it is not necessary. The act of straining itself can contribute to the sauce breaking.

Below is the picture of the sous vide version after I removed it from the bath. This is the THINNEST that veloute should ever be. At this thickness, the sauce is the most difficult to bring to effect, regardless of method. It also likes to scorch, something that sous vide will not do.

Note the speck of parsley in the eastern corner of the spoon. No matter what I do, a little parsley always manages to show up. This will be important later.

Below is the veloute that we prepared in the pan. It is exactly the same thickness, and if it was any thinner, there wouldn’t even be enough flour to hold it all together.

Just in case you thought I was trying to pull a fast one, you can clearly see that the parsley is now in Idaho. Obviously, a different example. If we can’t trust ach other now, all is lost, right?

Okay, NOW what?


Note: Midstream as we are, let’s make a quick adjustment. Here’s the formula that would be used for a MEDIUM thickness veloute:

  • Flour, 3 Tablespoons (instead of 2)/24g (instead of 18).
  • Bouillon, COLD, 2 cups/450ml.
  • Butter, 2 Tablespoons/30g.


As if this wasn’t confusing enough, I combined the two sauces, and added

Flour, 32 grams

to one third of that total mixture.

Hopefully, this will make a THICK veloute.

I mixed lightly to the paste stage, added the rest of the sauce, and packaged it in a Ziploc Gallon Freezer bag, as seen below:

I processed another hour. The sauce never separated into the two components like it did the first time, and as it turns out, we learned something. The more flour in the recipe at onset, the less the sauce will settle during sv processing. In other words, the thick stuff will not sink so much, and the thin stuff will not float so much.

Intuitively, this makes perfect sense to my 10K brain, but I don’t know how to explain it. I’ve done several versions of this over the last few days. In between contemplating what I’m going to do with all these creamy sauces, (green bean casserole, mac’n cheese), I noticed that the more flour I utilized, the less I had to tinker with the sauce once it came out of the bath. This will be clearly evidenced in the forthcoming installment covering béchamel.

And, after the allotted time, I poured this out of the bag into the container. No mixing.  It is somewhere between perfectly smooth and not completely perfectly smooth, which is the way it would come out if you whipped the living daylights out of it, anyway. More on that later!

Looks good from here!

Sure, I see them. I’m sure you see them too! Relax. By now, it is okay to strain the sauce, as long as your sieve is not too fine. And, when your done, look in the strainer and see how much was left behind. Not enough to detect other than by eye. I do not call these lumps. These are more like clusters. Clusters of balloons. MORE ON THAT LATER.

At service, just enough cream is added to veloute to cloud it. After all that effort, be sure not to boil the sauce after you add the cream. You won’t get lumps, but there’s a good chance you’ll get dots. I hate dots.

Below, I placed a half a crouton next to a ribbon of Licorice Gastrique, just for contrast. Nope, that’s not parsley. That’s chive.

When does school become OLD school?

I was talking to a friend of mine. We were discussing the various benefits and drawbacks of modern kitchen tools. Specifically, food processors, blenders, and stick blenders. At one point in the conversation about custard recipes, she said “for that, I still like the old school method–a wire whisk.” Well, yes, a wire whisk would be “older” school than those other devices. For better or for worse, I am old enough to remember when wire whisks were considered NEW school, even though new school and old school were not even terms in common use. Ah, the 70’s!

It’s true, though. I trained under chefs who insisted that apprentices should not be allowed to use wire whisks until they had mastered making sauces without them. The wooden spoon was the implement to be used. Now, that really IS old school. Not only custards, but cream sauces, veloute, hollandaise. We interpreted this training model as a form of punishment for our youthful folly, and none of those chefs contradicted it. If we dared to ask those guys “why,” well, most of us knew better. It was hard enough making sauces with a wooden spoon, without attracting the ire of a chef who had that sort of “old school” bent. Talk about poking the bear.

Well, I’ll be darned

There really was a reason. Unless informed otherwise, most people assume that when flour is introduced into a liquid, the flour dissolves. This creates thickness. Pretty basic, right? But this is not true. Flour does not dissolve, any more than pasta dissolves when cooked properly. Notice, I said “properly.” The flour itself is like tiny, individual, nano-grains of pasta. Little spheres, that, when exposed to water and heat, inflate and expand. The water that was outside, is now inside. Allegorically, it sort of reminds me of the way tapioca behaves. Great, now I have to write an article on tapioca.

Once the tiny balloons have inflated, all is well. You now have a sauce, and it appears thick because the water is inside the balloons, and there is only enough liquid outside to insulate the balloons from each other. To a greater or lesser degree.

That’s why you don’t want to whip a sauce. The friction caused by the whip, or the blender, will shred the balloons like so much mylar. Especially after they’ve been in the engorged state for a while. For those who attempt to reheat veloute on the day after, this should come as a revelation of sorts. It seems thick, but once you start heating and beating, the balloons all break. Attempting to strain the sauce also creates enough friction to damage the balloons.

and here’s the kicker.

Lumps, no lumps, what’s a lump, we jump through hoops in dread of a lump. But in the case of a veloute, as long as the flour was originally dispersed in the liquid, you are not going to bite into a lump of raw flour. Lumps in veloute may infuriate the chef, but they are almost impossible to detect in the mouth. And that doesn’t matter either. Why?

I can’t remember EVER seeing a veloute applied in a manner that would give some one the opportunity to detect lumps, whether they were there not. Veloute is (or was, really, you hardly see it any more) used to hold the lobster and mushrooms all together in a Lobster Thermidor. It’s used to hold the chicken together in a Fricassee,  and it’s used to barely coat the veal in a Blanquette. Good luck finding a lump in there, under all that parsley. There’s always somebody who has an excessively tidy sock drawer than can find a lump in the Shrimp Newburg, but for most of us, it’s just not that much of a concern.

Stay tuned for Béchamel!