Log in

No account? Create an account
Previous Entry Share Next Entry
Okay, so how *does* a pot pie gravy work?
Having made a leg of lamb the other day, I was looking for a good dish for the leftovers. So I decided to try a variation of the usual answer to turkey: pot pie.

The theory all seemed sound, and I was pretty confident of my flavor profile. I cut the leftover lamb into bite-sized chunks, and browned it in butter with pepper, coriander and a bit of salt. (The combination from Fidaush, one of my favorite period lamb dishes.) Then I made a roux out of boiled-down lamb jello and milk, thickened with flour over low heat until it was a reasonably good-looking white gravy. Put the meat and some frozen veggies into the pot, pour the gravy over, cover with a (commercial) crust, and bake until the crust starts to brown.

The flavors were good, and worked as planned. (A little too strong on the pepper, but that's a common sin of mine.) But the gravy completely failed -- what went in as a nice smooth white gravy completely broke in the pot, so I wound up with essentially lamb broth. The flour seems to have just kind of disappeared, as (mysteriously) did the milk.

So I'm mystified. I don't make rouxs often, so I presumably made a mistake, likely an elementary one, but I have no idea what. Are there traps I should be aware of when making a roux gravy to bake like this?

  • 1
The veggies were still frozen when going in? I think that's probably where the extra water came from. Your plan should have worked fine otherwise, not that I've made that many such pies, but I think it was sound.

Ah -- sure, that didn't occur to me, but it's a plausible theory. The veggies didn't *seem* wet because they were frozen, but could well have been carrying a lot of water. Should have gone with my instincts and given them a quick saute, which would have given them a chance to dry a bit.

(Or simply gone with fresh veggies, but it's hard to put together a good *mix* of fresh veggies in small quantity from the supermarket. When I'm shopping at Whole Foods, I mostly take my veggies from the salad bar, but that doesn't work as well at Stop and Shop...)

You'd be surprised how much water those things can carry. Even heating them in the microwave until steaming can get rid of a lot of that water, and is relatively quick.

Using the frozen veggie mixes is a good thing, and as long as you defrost, you can do a lot with them that would be a real pain to stock in the fridge (and find they had rotted after a week, etc, yuck.) Frozen can be very useful!

Yeah, that was my theory. But in practice, I've used frozen veggies very little, and don't yet quite know how to work with them correctly...

“Then I made a roux out of boiled-down lamb jello and milk, thickened with flour over low heat until it was a reasonably good-looking white gravy.”

What did you use for fat? A roux is mix of flour and fat (typically butter), generally cooked by themselves long enough for the flour to mellow a bit before any liquid is introduced. A lamb jello is a collagen rich broth, but not necessarily fatty. Whole milk is only about 4% butter-fat.

When I make turkey pot pie, I add flour aromatics that have been cooked in butter, and stir that for a while to absorb the butter before I add further liquid. I also heat up the whole filling on the stovetop before it gets a crust and goes in the oven. That way, I can adjust seasonings and thickness as needed, and the extra liquid from the frozen veggies gets stirred in.

Ah -- that could be another major problem. I included the lamb fat that had separated out of the jello, but that might easily have not been enough for the purpose.

Consulting Jane (who usually makes the roux), she confirms that she normally starts with cooking a bunch of butter and flour, before adding any of the liquids, just as you suggest. So it may be that my roux was fatally flawed from the beginning, the frozen veggies aside.

Oh, well -- live and learn. At least the flavor proved itself out, so I think I'm on the way towards a good recipe once I work out these snags...

Yep, that's right. A roux is classically equal parts fat and flour, and is cooked alone in the pan before you add the liquid. (The lamb jello is "liquid" for this purpose.) Once the roux is cooked, the liquid of choice is whisked in, a little at a time so you don't get lumps.

Degree of cooking determines thickening capacity. A white roux is cooked for relatively little time, and contributes little flavoring but lots of thickening power. Red and black roux are more flavorful but do not thicken as much.

To add or elaborate:
One of the major purposes of a roux in sauce-making is a way to convey flour to the liquid without clumping. A properly-made roux will, in theory, encapsulate the flour granules in fat, sort of like an emulsification. That way each particle can absorb its but of liquid and thicken the sauce without adhering to its neighbor. As I understand it, as the roux darkens this thickening property is reduced somewhat but, with the Maillard reaction you get nice toasty flavors and colors (in grain you get a somewhat different trade-off - gaining color and flavor while losing enzymes and sugar potential; in coffee you lose caffeine while gaining the roast).

It may be that the frozen veggies gave you some extra liquid, and it may be that the flour wasn't "emulsified" properly/enough and clumped to itself (or something else) without being separate enough to thicken. My personal experience has been that you'd not get broth with just too much liquid, but rather a thin gravy; if you're getting broth the flour didn't get to dancing at all (hey - wall-flour!).

Okay, that tracks. The core of my mistake seems to have been misunderstanding the mechanics of how roux works. Not too worried on the flavor front -- the point here is mainly the flavor from the lamb jello itself -- I just managed to create something that *looked* like a white roux at first (and had the right consistency until it broke) but didn't behave like one...

I've found that heating the liquid (very warm to the touch, but not burning) before I add it reduces the number of lumps that I then have to whisk back out.

(Deleted comment)
Cool -- this is all consistent with what I'm hearing from cvirtue and oakleaf_mirror, so I seem to be getting a clearer idea of how it should work. All fixable, so I'll have to find an excuse to try this again sometime soon, to get a recipe worth writing up.

(And next time, I really should measure my pepper and coriander, so I can adjust the seasoning for people who don't like things as peppery as me. Cracking fresh pepper directly into the dish gets great flavor, but makes it a bit trickier for me to fix the details...)

A belated comment on this post:

You do need to start out with equal parts fat & flour in a roux but the fat doesn't necessarily have to be butter. Most fats will work though saturated fats, being more solid at room temperature, will generally work better. I often use the fat from pan drippings to make gravy but I do it after I've deglazed the pan and separated it out from the liquid/gelatinous portions. Then I take some of the separated fat and mix it with flour to make the roux stirring in the de-fatted drippings once the roux has cooked to the desired amount.

Most cooks know the easiest way to extract the rendered fat by chilling the drippings enough that the fat solidifies enough to be pulled off the still liquid/gelatinous drippings below. This isn't exactly practical when making a gravy to serve with the just roasted meat however. For that, there are gravy separator cups widgets that let the drippings settle out and be poured off from the insoluble liquid fat. My experience has been that the usual type on on the market work poorly. They rely on a channel running from the bottom of the cup to a top spout through which the dripping liquids can be poured out from underneath the lighter weight insoluble fats. It's not a very good design; since the fat is on top it often sloshes out as well when pouring out the drippings. Making the top of the cup wider so the fat layer is narrower only makes it unwieldy to handle.

When I worked in a chem lab we had separatory funnels to separate insoluble liquids. I've wondered for years why they didn't make a kitchenware version of it. Finally someone has. My only quibble is that a more funnel shaped bottom like the lab ware version would be nice but it's not a real problem. I used it at Christmas on the turkey drippings and it worked like a charm.

One caveat: the valve at the bottom held OK but I'd still keep it over a saucer for drips. Swing-A-Way products are usually reasonably well made and this is no exception. It's not flimsy but then again it's not exactly a high-tech precision tool either. The cost of manufacturing an absolutely drip-tight seal would push the sales price too high for the market at the cost margin they're looking at. I recommend acquiring one even if you only make gravy a couple of times a year, it saves a lot of hassle and gives you good control over the amount of fat in your gravy so it never comes out too greasy.

Oooh, that's a nifty toy. I may have to get one of those...

  • 1