Having spent money on the vacuum pump, I was reluctant to spend still more on a vacuum chamber. Conceptually, it's enormously simple; the only hard bit is collecting all the parts. Basically, it's just a twelve-inch piece of 6-inch diameter schedule 40 PVC pipe, plus a half-inch of acrylic epoxied to one end.
I managed to get the belled-out end when the guy at the hardware store cut it off. As it turns out, this is just big enough to allow a bleach bottle to fit. The picture here is of a different, straight, section.

Wrap a string or some paper or something around the ends so you can get a line to cut to that's more or less even all the way around. Once you've cut it, work the end around on a piece of sandpaper to flatten it out.

This is a closeup of the bottom of the chamber. I cut the acrylic myself, rather than pay for the plastic store to cut it for me. TAP plastics, if you're in the Silicon Valley area has a bin of cut-offs you can rummage through.

As you can see, I didn't manage to keep all the air out of the surfaces between the two plates. Two 1/4 inch plates cemented together has held up just fine through many cycles. I have no conception whether this is overkill or not.

Another piece of acrylic forms a very flat ledge that forms part of the seal. As with the bottom, it's epoxied in place.
More acrylic sandwiched together makes the lift-off top for the chamber. This is a bit under an inch thick; probably more like 2cm. As before, I have no idea if this is overkill, gross overkill, or riding the ragged edge of disaster. On the other hand, it hasn't failed in well over a hundred cycles, and it doesn't show any signs of cracking or strain, either.
A hole, a brass hose barb, some epoxy, plus automotive rubber hose penetrates the centre of the top. In this case, as best I recall, it's 1/4 MPT, 3/8 i.d. hose barb, and 9/16 inch rubber i.d. hose.
A rubber mat from American Science & Surplus helps form the seal. Do not skip this. Acrylic is pretty flat, but this element allows you to be less than perfect in your craftsmanship, and also is removable for cleaning when plaster or alginate or whatever splatters on it.

My previous setup inverted this whole idea; there was a solid base, plus a trap to catch any spillover. I kept worrying about the trap (a glass spaghetti sauce bottle) failing suddenly, even in its PVC shroud, so I reorganized the setup not to need a trap. So far, I've had a few spills, but nothing so huge as to actually get crud into the vacuum hose.

Here, one valve for each vacuum pump, plus one valve to the atmosphere to break the suction. A vacuum gauge from the hardware store completes the manifold.

Beware: of about 7 vacuum gauges at the local hardware store, only one even started with the needle anywhere near zero, and it claims that cold water boils around 27 inches of mercury. This is complete and utter nonsense, but it does a slightly better job of indicating where things are than just listening to how hard the pumps are laboring.

It turns out that the low-vacuum vacuum pump's seals are so bad that I need to actually use a hand valve to take it out of the system once the good pump really gets going.

The whole setup. The Welch vacuum pump is on the left, while the low-vacuum roughing pump is on the right. They're both plugged in to the circuit box in between them. Each has its own switch. Hoses run from the pumps to the manifold, and thence to the top plate.

Technically, I'm probably ruining the Welch by putting so much water through it, but I change the oil pretty religiously. The nice people at Duniway Stockroom Corp (who sold me an unreconditioned return very cheaply) say that phenolic vanes in a vacuum pump will sieze up, but that steel vanes may well survive this treatment. Time will tell.